Combining Implicit Measures And Information Foraging Theory To Improve Web Search

The world wide application of search engine can be a milestone of the evolution of Internet. The increase number of people is willing to choose search engine as the primary tool for hunting resoures from Internet. But the performance of search engine is not always satisfactory. The massive distributed, dynamical and evolutionary characteristics of World Wide Web inspire us study it drawing on the Information Foraging Theory which assumes that people prefer yield more useful information per unit cost. Understanding the value of implicit measures is also important to help World Wide Web users search more effectively.Borrowing idea from information foraging theory we propose a method to estimate the Web page information gain based on the implicit measures analysis. This thesis fiestly developed an experimental search platform to record the URLs of the pages be browsed, the time spent reading, the time spent scrolling and the number of links be clicked. Secondly the thesis analyzed these data using multiple linear regression modeling and obtained a regression function which could be used in calculating the information profitability of the Web pages. And then we regrouped the Web pages searched out by the descending order of the Web page information profitability. We also reported experimental data to show that the search results be regrouped can increase searching efficiency. At last,the findings suggest that the combination of information foraging theory and implicit measures analysis support effective Web-search interaction for everyone.